An autotransformer is an electrical transformer with a single continuous winding that shares a common magnetic core and electrical connection to act as both the primary and secondary coil. Unlike a traditional dual-winding isolation transformer that transfers energy purely through magnetic induction across separate coils, an autotransformer passes a significant portion of its load current directly through the electrical connection of the shared winding. This fundamentally changes a real circuit by allowing you to step voltage up or down using drastically less copper, core steel, and physical volume, but it entirely eliminates galvanic isolation between the source and the load. Because of this shared electrical path, people commonly confuse autotransformers with isolation transformers—a dangerous misconception that leads hobbyists to believe plugging a device into a Variac provides shock protection. It does not.
To understand the massive physical advantage of this topology, look at how a standard 50 kVA isolation transformer compares to an autotransformer handling the exact same 50 kVA load when stepping 480V down to 400V.
| Parameter | 50 kVA Dual-Winding Isolation | 50 kVA Autotransformer |
|---|---|---|
| Physical Core/Winding Rating | 50 kVA | 8.3 kVA |
| Copper Weight (Approx) | ~90 lbs | ~15 lbs |
| Efficiency at Full Load | 97.5% | 99.2% |
| Short-Circuit Impedance | ~4.5% | ~0.8% (Higher fault current) |
| Galvanic Isolation | Yes | No |
The Core Math: Why Autotransformers Are Smaller and Cheaper
The reason an autotransformer can be a fraction of the size of an isolation transformer comes down to the difference between conducted VA and transformed VA. In an autotransformer, the load current flows directly from the input line through the common winding to the output. Only the difference in voltage between the primary and secondary actually has to be transferred via magnetic induction.
Let us run a worked numeric example. Suppose you have a 50 kVA industrial load that requires 400V, but your facility supply is 480V. You need a step-down transformer.
The physical size of the autotransformer is determined only by the 'transformed' portion of the power, calculated using the co-power ratio formula:
S_transformed = S_load × (1 - (V_low / V_high))
Plugging in our real-world values:
- S_load = 50,000 VA
- V_low = 400V
- V_high = 480V
S_transformed = 50,000 × (1 - (400 / 480))
S_transformed = 50,000 × (1 - 0.833)
S_transformed = 50,000 × 0.167 = 8,350 VA (8.35 kVA)
However, this mathematical advantage shrinks as the voltage ratio increases. If you were stepping 480V down to 120V, the transformed VA would be much closer to the total load VA, and the size/cost savings would diminish significantly. For ratios greater than 3:1, dual-winding isolation transformers usually become the more practical and safer choice.
Where You Meet Autotransformers in Practice
You are likely already using autotransformers on your bench or in your facility, even if you do not call them by their formal name.
1. Variable AC Supplies (Variacs)
The classic Staco Variac is a variable autotransformer. A single toroidal winding is wound on a core, and the top layer of enamel insulation is machined off to expose bare copper. A carbon graphite brush rides on this bare track, allowing you to smoothly dial the output from 0V to slightly above line voltage (typically 0-140V on a 120V input). Because it is an autotransformer, the output is directly referenced to the input neutral. Never assume a Variac output is safe to touch; it is electrically live to the mains.
2. HVAC Buck-Boost Transformers
If you install a 208V commercial AC compressor in a building with a 240V supply, you need to 'buck' the voltage down by 32V. Instead of buying a massive 10 kVA isolation transformer, HVAC technicians use a small 2 kVA buck-boost transformer wired as an autotransformer. By wiring the 16V secondary in series-opposing with the 240V primary, the output drops to 224V (or 208V depending on the exact tap). A $150 buck-boost unit can easily handle a 30A, 8 kVA compressor load because it is only transforming the 32V difference.
3. Reduced-Voltage Motor Starters
Large industrial induction motors draw massive inrush currents. Korndörfer starters use a three-phase autotransformer to temporarily feed the motor at 50% or 65% voltage during startup, reducing the mechanical shock and electrical grid sag, before bypassing the transformer entirely for run-mode.
The Fatal Flaw: Loss of Galvanic Isolation
The single biggest hazard when working with autotransformers is the false sense of security they provide. In a dual-winding isolation transformer, the secondary circuit is floating. If you touch one secondary wire while standing on the ground, you will not complete a circuit back to the primary source.
In an autotransformer, the input and output share a common physical wire. If you use a 240V-to-120V step-down autotransformer and the common neutral connection breaks or is wired incorrectly, the '120V' output terminal can instantly rise to the full 240V line potential. Furthermore, because the neutral is shared, a fault on the load side can backfeed directly into the mains without the magnetic gap that limits fault currents in isolation transformers.
FAQ: Common Bench and Jobsite Questions
Can I use a standard buck-boost transformer as an isolation transformer?
Yes, but only if you wire it as one. A standard buck-boost transformer has two separate primary coils and two separate secondary coils. If you wire the primary coils to the source and only use the secondary coils for your load (leaving the primary and secondary entirely unconnected to each other), it acts as an isolation transformer. However, its kVA rating will drop to its nameplate series rating (e.g., 2 kVA), not its autotransformer boosted rating.
Why do Variable Frequency Drives (VFDs) sometimes require isolation transformers instead of autotransformers?
VFDs generate high-frequency common-mode noise and voltage spikes due to rapid IGBT switching. An autotransformer's low impedance and direct electrical connection will pass this high-frequency noise straight back onto the facility mains, potentially corrupting sensitive PLCs or communication buses. A dual-winding isolation transformer with an electrostatic Faraday shield blocks this common-mode noise, protecting the rest of the facility.
How do I size the overcurrent protection for the common winding?
The common winding only carries the difference between the load current and the input current. However, standard protection practices dictate that you must size the primary overcurrent device based on the full input current of the autotransformer, not just the transformed current, to ensure the entire magnetic circuit is protected against saturation and thermal overload during fault conditions.
